1. When might a non-parametric test be more appropriate than a t-test or ANOVA?

Non-parametric tests are more appropriate when the data violate key assumptions required by parametric tests, such as normality or equal variances. They are particularly useful for ordinal data or when sample sizes are small and it’s difficult to confirm normal distribution. In these situations, non-parametric tests provide a reliable alternative that does not rely on strict distributional assumptions.

2. How do non-parametric results differ in interpretation (e.g., medians vs. means)?

Unlike parametric tests that compare means, non-parametric tests typically compare medians or use the ranks of the data. This means the results focus on the central tendency or overall distribution rather than average values. Interpreting non-parametric results involves understanding whether one group tends to have higher or lower scores relative to another, rather than the precise magnitude of difference between group means.

3. What are the trade-offs between parametric power and non-parametric robustness?

Parametric tests generally offer greater statistical power when their assumptions are met, allowing them to detect smaller effects with fewer participants. However, they are sensitive to violations of those assumptions, which can compromise results. Non-parametric tests are more robust and less sensitive to such violations, but this robustness comes at the cost of reduced power, meaning they might require larger sample sizes to detect significant differences.

Final Write-up — Results & Discussion

Results

A paired-samples t-test was conducted to examine whether students’ scores improved from the pre-test to the post-test. Results showed a significant increase in scores from the pre-test (M = 60.7, SD = 3.56) to the post-test (M = 68.9, SD = 4.24), t(9) = 20.43, p < .001. The effect size was very large (Cohen’s d = 6.46), indicating that the intervention had a strong impact on student performance. Overall, this suggests that students performed significantly better after the study period.

More so, to explore whether gender played a role in post-test scores, an independent-samples t-test was run. The results showed that female students (M = 70.0, SD = 1.63) scored slightly higher than male students (M = 65.6, SD = 3.36). However, this difference was not statistically significant, t(8) = 1.74, p = .121. The effect size was small to moderate (Cohen’s d = 0.94), but the results suggest that there was no clear gender-based difference in performance.

Lastly, a one-way ANOVA was conducted to test if the study method had a significant effect on post-test performance. The results were significant, F(2, 7) = 6.52, p = .025, with a large effect size (η² = .65), indicating that study method influenced how well students performed. More so, Tukey’s HSD post hoc test showed that students who used flashcards (M = 71.0) and group study (M = 70.0) scored significantly higher than those who only used reading (M = 61.7). However, there was no significant difference between the flashcards and group study methods. Therefore, this suggests that both were equally effective and better than reading alone.

Discussion

The results clearly show that students improved after the intervention, which reflects well on the study activities or preparation that were used during the period. The large effect size from the paired-samples t-test means the improvement was not just by chance; it had a real impact. While females had slightly higher post-test scores than males, the difference was not significant. This suggests that both genders benefited fairly equally from the learning process.

The most important finding was the influence of the study method on the student’s performance. Students who used flashcards or engaged in group study performed significantly better than those who just read the material. This, in turn, reinforced the value of active learning strategies. Flashcards promote active recall, while group study encourages discussion and explanation — both of which are supported by cognitive psychology as more effective than passive learning (Sternberg & Sternberg, 2016). Although flashcards and group study were equally effective here, either one offered a much better learning experience than reading alone.

From a practical point of view, these results have clear implications for how students should be encouraged to study. Teachers, tutors, and learning support services should help students move away from passive methods like reading and instead teach them how to use tools like flashcards or how to form productive study groups. These strategies do not require expensive resources but can make a noticeable difference in learning outcomes. Overall, this study adds to the growing evidence that how students study matters just as much – if not more – than how long they study.

References

Sternberg, R. J., & Sternberg, K. (2016). Cognitive psychology (7th ed.). Cengage Learning.